Enforce shorter AI responses and prevent cut-off sentences

- Reduce max_tokens from 100 to 75 for shorter output
- Add truncate_to_complete_sentence() to trim at last punctuation
- Applied to both chat and auto-respond paths

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

backend/services/llm.py

@@ -124,7 +124,7 @@ class LLMService:
                         json={
                             "model": self.openrouter_model,
                             "messages": messages,
-                            "max_tokens": 100,
+                            "max_tokens": 75,
                         },
                     )
                     response.raise_for_status()